Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Attribute-level sentiment analysis method and system based on three-dimensional target matrix

A technology of three-dimensional target and sentiment analysis, applied in the field of data processing, can solve problems such as low overall efficiency and inability to make full use of information, and achieve the effect of good performance

Pending Publication Date: 2022-05-10
SHANDONG EVAYINFO TECH CO LTD
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The above method only performs simple interactive operations on the two related tasks of star classification task and attribute-level sentiment classification on the shared coding layer, and cannot make full use of the respective information in the original two tasks. In addition, it needs to decode the comment content multiple times. The star attribute and all classification attributes are less efficient overall

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Attribute-level sentiment analysis method and system based on three-dimensional target matrix

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0044] This embodiment discloses an attribute-level sentiment analysis method based on a three-dimensional target matrix. By using the three-dimensional information of the three-dimensional target matrix, the star rating, attributes, and emotional polarity are integrated to supervise the training and learning of the model, so that the model The correlation information between the two related tasks of star classification task and attribute-level sentiment classification can be comprehensively exploited.

[0045] See attached figure 1 As shown, in the specific steps:

[0046] Step 1: Process each comment sentence in the training set of the dataset into the input format required by the BERT pre-training model. Each sentence is followed by 19 attributes, including a star attribute (1 to 5 stars), indicating There are also 18 tags classified by attribute for the overall emotional point of view of the comment sentence, which are: dish-portion, dish taste, environment-decoration, lo...

Embodiment 2

[0067] The purpose of this embodiment is to provide a computing device, including a memory, a processor, and a computer program stored in the memory and operable on the processor, and the processor implements the steps of the above method when executing the program.

Embodiment 3

[0069] The purpose of this embodiment is to provide a computer-readable storage medium.

[0070] A computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the steps of the above-mentioned method are executed.

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides an attribute-level sentiment analysis method and system based on a three-dimensional target matrix, and belongs to the technical field of data processing, and the method comprises the steps: processing each comment sentence in a training set into a required input format; on the basis of the format, constructing a three-dimensional target matrix capable of representing all attributes of the corresponding comment sentences at a time; inputting the processed training set into a model for supervised learning training, obtaining a first matrix vector representation at the last layer of the model, inputting the first matrix vector representation into a pooling collection layer to obtain a second matrix vector representation, and obtaining a final third matrix vector representation through a linear transformation layer and activation operation; every time the model is trained, the trained model is verified on the verification set, and an optimal model is obtained; inputting comment sentences to be analyzed into the optimal model, and decoding the sentiment polarity representing each attribute.

Description

technical field [0001] The invention belongs to the technical field of data processing, and in particular relates to an attribute-level emotion analysis method and system based on a three-dimensional target matrix. Background technique [0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art. [0003] With the rapid popularization and rapid development of the Internet, hundreds of millions of users obtain information and express their opinions and emotions on the Internet. [0004] In the field of e-commerce, sentiment analysis of user comments can provide references for consumers to purchase goods and make decisions about store selection, and can also help merchants to better understand user suggestions and needs, thereby improving merchant services. quality and level. [0005] In attribute-level sentiment analysis involving two related tasks, the previous method designed the...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06F40/30G06F16/35G06K9/62
CPCG06F40/30G06F16/35G06F18/217G06F18/214
Inventor 李钊赵秀浩辛国茂陈通吴士伟孙浩宫传华
Owner SHANDONG EVAYINFO TECH CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com